PHILADELPHIA – Tucked away in private partnering rooms, far from the hustle and bustle of exhibitors, panel discussions and keynote luncheons, executives conducted 29,279 meetings during the 2015 BIO International Convention. For Roche AG, "there are three general areas that we're looking at," said Jason Coloma, global head of venture and innovation, and soon-to-be head of oncology. Roche, he said, wants complementary assets for its pipeline, diagnostic and biomarker tools that will build on Foundation Medicine Inc.'s biomarker development and clinical trial collaborations to leverage its pipeline. Beside Coloma with her own "want list" in mind sat Sophie Kornowski-Bonnet, head of Roche Partnering, who is focused on neurodegenerative and rare diseases.

The two visited with roughly 100 potential partners during their time in Philadelphia.

"It's not the number, it's the quality," Kornowski-Bonnet told BioWorld Today.

Kornowski-Bonnet has worked in partnering for four years and has learned quickly that "external innovation is a must-have." Not everything can be developed internally, as it was in the early years.

"We cannot have it right completely. Over half of the innovation comes from external seeds," which include ideas direct from academia and biotech, she said. "All of us nowadays have beefed up our partnering capabilities."

Approximately one-third of Basel, Switzerland-based Roche's pipeline stems from partnering activity, and it currently is developing about 60 products in the pipeline. In addition, 37 percent of the company's sales come from partnered products.

"Some studies show that it will expand over time," Kornowski-Bonnet said. "What is clear is that it's extremely important for an organization" to work outside its own researchers to find companies working on specific targets of interest.

Coloma said that partnering for oncology products has become "increasingly competitive" over time because it is a good growth driver. "As a result of that, you're seeing more deals being done in the preclinical phase than in any other therapeutic areas."

Kornowski-Bonnet created the venture and innovation division to focus on early access and to engage with companies before the crowd comes.

"Deals come in many shapes and forms," she said. "It depends on what we are keen" to do, and what the risk entails. "Sometimes you can do licensing. Sometimes you are going to co-invest. You can do an option deal and bring in innovation later on. Sometimes it's a straightforward acquisition."

DECIDING WHEN TO POUNCE

The timing of forming partnerships is a tricky business and mistakes are easily made.

"There is the question of, 'is the science sufficiently proven,'" Kornowski-Bonnet said. "We are really trying to go into proven science, but at the same time we try to be there where it's early enough but not yet crowded."

Roche holds internal discussions and brings in advisory boards to ascertain what competitors are doing, "so we invest at the moment that the science is ripe," she added.

There are some lessons learned, as is the case of siRNA, she said: "We thought we were there, but in terms of delivery we weren't."

Several pharmaceutical companies entered the RNAi space in the mid-2000s, but the majority of pharma firms have backed away, leaving the assets in biotech hands. Roche is no exception, having exited or abandoned eight programs over the past decade.

Coloma said they stay close to the research organization and define a strategic want list "all the way down to the mechanism of action."

In April, for instance, Roche gained later-stage assets through New Delhi, India-based Curadev Pharma Private Ltd., including IDO1 (indoleamine-2, 3-dioxygenase-1) and TDO (tryptophan-2, 3-dioxygenase) inhibitors, enzymes that mediate cancer-induced immune suppression. The research collaboration, which includes a preclinical lead compound, entailed a $25 million up-front payment for Curadev, as well as up to $530 million in milestone payments, and escalating royalties that could reach double digits for the first product.

Kornowski-Bonnet said, "Sometimes you wait for the science to be strong enough so you can truly follow them."

For Alzheimer's disease, Roche is focused on various targets, including Tau (RG7345), monoamine oxidase-B (MAO-B; RG1577), and amyloid beta (Abeta; RG7412, crenezumab; and RG1450, gantenerumab), but doesn't truly know which pathway will lead to successful development of a drug. The Alzheimer's candidates in phase II are RG1577 partnered with Evotec AG, of Hamburg, Germany, and crenezumab partnered with AC Immune, of Lausanne, Switzerland, and in phase III is gantenerumab partnered with Morphosys AG, of Martinsried, Germany.

"You have to get the biology right, but you also have to get the clinical trial right," Kornowski-Bonnet said. "It's a very complex field. The Abeta hypothesis is starting to emerge as probably one that will bear fruit."

"If you look at cancer immunology," Coloma added, "our basic understanding of the biology was pretty limited." Although the checkpoint inhibitors PDL1, which inhibits activated T cells, and CTLA4, a blocker of the costimulatory system, are validated, many questions remain, such as the combinations and sequence of combinations that should be used.

Roche is looking not only for the right therapeutics to license, but is strongly focused on biomarker development. At the 33rd Annual J.P. Morgan Healthcare Conference in San Francisco in January, Roche announced its $1.05 billion acquisition of majority ownership of Cambridge, Mass.-based Foundation Medicine, which holds comprehensive genomic profiling products. (See BioWorld Today, Jan. 13, 2015.)

SUCCESSFUL DEALS

In retrospect, the most successful deals become obvious, although the challenge is to repeat them in the future.

Kornowski-Bonnet cited the 2005 acquisition of Glycart Biotechnology AG's glycol engineered antibodies that later turned into Gazyva as a prime example. "Whenever you can go from innovation into making a difference in prolonging lives, I think it's successful," she said.

Roche acquired Glycart, of Schlieren, Switzerland, for CHF235 million (US$180 million) in cash. In 2013, phase III data of lead product GA101 (obinutuzumab), a glycol-engineered, Type II, humanized anti-CD20 monoclonal antibody, showed it bested Rituxan (infliximab, Roche AG and Biogen Inc.), in previously untreated chronic lymphocytic leukemia patients. It gained FDA clearance as Gazyva under the breakthrough-therapy designation later that year. Phase III data released earlier this year in the indolent form of non-Hodgkin's lymphoma showed Gazyva reached its primary endpoint of progression-free survival early, with data in May showing it reduced the risk of disease worsening or death by 45 percent when used with bendamustine, compared with bendamustine alone. (See BioWorld Today, July 25, 2013, Nov. 4, 2013, and Feb. 5, 2015.)

Coloma agreed that the most successful deal was securing Gazyva, in which its executive Pablo Umana remained with Roche, heading up early stage cancer immunology efforts. The acquisition brought Roche engineering capabilities that Umana built and provided a differentiated approach to cancer immunology. The deal "to secure an anti-CD20 molecule has evolved into a center of excellence in terms of protein engineering," Coloma said.

Last August's acquisition of Brisbane, Calif.-based Intermune Inc. for $8.3 billion also came to mind when discussing successful deals. The lead product, Esbriet (pirfenidone), for idiopathic pulmonary fibrosis, inhibits the synthesis of TGF-beta and TNF-alpha. Upon acquisition, it was already approved in Europe, Canada and Asia, but gained FDA approval a few months later in October. Roche priced it at $94,000 per year in the U.S. (See BioWorld Today, Aug. 26, 2014, and Oct. 17, 2014.)

"At the end of the day, the clinical trial they did was able to demonstrate true mortality benefit in a very severe disease where patients had life expectancy of two to five years," Kornowski-Bonnet said.

"The drug is on a very good start," she added, citing promising feedback from prescribers and patients.